Cylindrical panoramas may be constructed using a single rotating camera. As the camera is rotated, images may be captured at defined increments until the desired panoramic field of view has been traversed. Vertical strips may then be extracted from the center of each image, and the strips can be placed next to one another to form a single uninterrupted cylindrical panoramic image.
This process can be extended to create cylindrical stereoscopic (e.g., three-dimensional) panoramic images. For example, two cameras can be mounted, one next to the other, separated by a defined distance. The cameras may then be rotated in unison about a point halfway between them. Each camera can be used to create a separate cylindrical panorama using concatenated vertical image slices, as described above. When the two resulting panoramas are viewed together, one by an observer's left eye and the other by the observer's right eye, a stereoscopic effect is achieved. However, while the rotating two-camera model may be useful for creating still stereoscopic images, the system described does not lend itself to efficiently providing a moving stereoscopic panoramic image.